CN107276093B - Power system probability load flow calculation method based on scene reduction - Google Patents
Power system probability load flow calculation method based on scene reduction Download PDFInfo
- Publication number
- CN107276093B CN107276093B CN201710552406.4A CN201710552406A CN107276093B CN 107276093 B CN107276093 B CN 107276093B CN 201710552406 A CN201710552406 A CN 201710552406A CN 107276093 B CN107276093 B CN 107276093B
- Authority
- CN
- China
- Prior art keywords
- scene
- initial
- probability
- collection
- cut down
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 57
- 238000009826 distribution Methods 0.000 claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000005315 distribution function Methods 0.000 claims description 8
- 230000035945 sensitivity Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 238000002347 injection Methods 0.000 claims description 5
- 239000007924 injection Substances 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 238000012546 transfer Methods 0.000 claims description 5
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 4
- 239000012141 concentrate Substances 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 230000000717 retained effect Effects 0.000 claims description 3
- 238000004146 energy storage Methods 0.000 abstract description 6
- 238000010187 selection method Methods 0.000 abstract description 2
- 238000012614 Monte-Carlo sampling Methods 0.000 abstract 1
- 238000012216 screening Methods 0.000 abstract 1
- 238000005457 optimization Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 239000000243 solution Substances 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- H02J3/383—
-
- H02J3/386—
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
- Y02E10/56—Power conversion systems, e.g. maximum power point trackers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a power system probability load flow calculation method based on scene reduction, which comprises the following steps: monte Carlo sampling is carried out on random factors such as a fan, a photovoltaic device, an energy storage device and a controllable load in an electric power system, the running state of the electric power system is obtained, and an initial scene library is constructed; reducing the initial scene library by using a rapid forward selection method, screening out typical scenes, completing the classification of the scenes and generating a scene set; on the basis, performing probability load flow calculation on each scene set by using a semi-invariant method to obtain the probability distribution characteristic of the state quantity of the power grid; and superposing the power flow characteristics of the scene sets, so as to obtain the overall power flow distribution characteristics of the power grid. The method provided by the invention can be applied to the load flow calculation of a large-scale complex power grid containing high-proportion renewable energy, can effectively solve the problem of the accuracy reduction of the load flow calculation result of a semi-invariant method caused by the large fluctuation of random factors of a power system, and ensures the high accuracy of the probability load flow calculation of the power grid.
Description
Technical field
The invention belongs to the Electric Power Network Planning fields of electric system, relate generally to the electric system of the renewable energy containing high proportion
Load flow calculation.
Background technique
Electric Power Network Planning is the important component of Power System Planning, work be according to the load growth of planning period and
Power source planning scheme determines best grid structure with meeting economic and reliable the needs of transmitting electric power.Load flow calculation is Electric Power Network Planning
Important content, main task is that the operating status of whole system is determined according to given service condition and network structure, is
The foundation of Power System Planning offer base reference and verification.
In recent years, with the big rule of the enchancement factors such as the new energy such as extensive wind energy, solar energy and energy storage, controllable burden
Mould is grid-connected, and a large amount of randomnesss and uncertainty occurs in electric power system source end, and the trend in power grid becomes probability by certainty trend
Property trend.This causes great interference to the accuracy of electric network swim calculated result, it is therefore necessary to containing at high proportion with
The tidal current computing method of the power grid of machine factor improves.
At present there are mainly three types of probability load flow calculation methods: using Monte Carlo Method as the simulation of representative, with point estimations
For the approximation method of representative and using Cumulants method as the analytic method of representative.Monte Carlo Method is the simulation of representative, is by setting
Determine random process, generate time series repeatedly, often calculation amount is very big in order to guarantee accuracy, causes time-consuming serious；Point estimation
Method is to seek the probabilistic method of each rank square of stochastic variable to be asked according to the probability distribution of known stochastic variable, opposite to cover spy
Caro method calculation amount is smaller；Cumulant is a numerical characteristic of stochastic variable, convolution sum deconvolution can be calculated abbreviation and be
The operation of the addition and subtraction of several cumulant, can be such that calculation amount substantially reduces.
Based on the analytic method of Cumulants method when calculating Probabilistic Load, calculating speed is fast and computational accuracy
Height, but Cumulants method is higher to data sensitivity.The Electric Power Network Planning of the renewable energy containing high proportion, power grid is in large scale,
Structure is complicated, and enchancement factor type is more, ratio is high, very huge by the number of scenes to raw data samples and discretization generation
Greatly, and since the randomness of enchancement factor (especially generation of electricity by new energy, such as wind-powered electricity generation, photovoltaic etc.), fluctuation are very big, therefore very
More scenes do not have representativeness or scene error itself is excessive, lead to not obtain accurate calculation of tidal current.In addition, adopting
During obtaining cumulant with statistical method, the boundary between sample is difficult to determine.
Scene in Electric Power Network Planning is exactly a kind of operation states of electric power system, and the generation of scene is exactly to obtain electric system
Operating status.So-called scene is cut down, and is exactly to cut down under the premise of not impact evaluation precision original scene, is rejected not
Scene representative, with error retains typical scene.Two classes, base are broadly divided into the technology that scene is cut down at present
In the reduction technology of distance and reduction technology based on density.Reduction based on distance develops comparative maturity, using comparing at present
Widely mainly there are synchronous back substitution method, fast forword back-and-forth method.Reduction technology based on density is mainly clustering methodology.Wherein
Fast forword back-and-forth method algorithm is intuitive, high-efficient.
Have not yet to see by scene abatement with Cumulants method ining conjunction with being directed to contain high proportion enchancement factor (blower,
Photovoltaic, energy storage device, controllable burden) power grid tidal current computing method.
Summary of the invention
The purpose of the present invention is to provide a kind of Probabilistic Load calculation methods cut down based on scene, to reduce
Caused by data fluctuations are larger when enchancement factor ratio is higher in power grid the problem of probabilistic load flow accuracy decline.
In order to achieve the above objectives, the invention adopts the following technical scheme:
1.1) enchancement factor in electric system is sampled, obtains initial scene library；
1.2) initial scene library is cut down, obtains typical scene；According to typical scene to complete in initial scene library
Portion's scene is classified, and scene collection is obtained, and scene collection is calculated according to the scene classification of initial scene library and probability characteristics
Probability characteristics；
1.3) probabilistic load flow is carried out to each scene collection using the Probabilistic Load Flow algorithm based on Cumulants method respectively；
1.4) probability characteristics for combining scene collection carry out probability superposition to the characteristics of tidal flow of each scene collection, obtain power grid
Trend distribution character.
The step 1.1) is specifically includes the following steps: obtain the feature of enchancement factor according to the initial data of electric system
Then distribution carries out n times sampling to enchancement factor using Monte Carlo Method, every progress single sample just obtains an initial scene
And its probability characteristics.
The step 1.2) specifically includes the following steps:
1) initial scene is cut down with fast forword back-and-forth method, until the initial scene retained reaches the target of setting
Scene number n, using the n of reservation initial scenes as typical scene；
2) classified according to the probability metrics between initial scene and typical scene to initial scene, made just by classification
Beginning scene is converged according to typical scene, so that it is determined that the boundary between initial scene；
3) the probability q occurred for j-th of scene collectionjIt is obtained by following formula:
Wherein, i ∈ j indicates that i-th of scene will be divided into j-th of scene collection, p in initial scene libraryiFor initial scene
The probability that i-th of scene occurs in library.
To Mr. Yu's initial fields scape oi, calculate separately the initial scene and each typical scene sjBetween probability metrics；It should
Initial scene is divided into one kind with the minimum corresponding typical scene of probability metrics, i.e., i-th of scene will be divided in initial scene library
To scene collectionpic(oi,sj) it is scene oiAnd sjBetween probability metrics, c (oi,sj)
For scene oiAnd sjBetween Euclidean distance.
The step 1.3) specifically includes the following steps:
1) by electric system modal equation and branch equation Taylor expansion is carried out at typical scene, then pass through line
Property, obtain sensitivity matrix and transfer matrix；
2) all scenes are concentrated for scene, the L rank central moment of scene are calculated using the method for statistics, according to the L
Rank central moment calculates the L rank cumulant of power grid injection variable, and L rank cumulant, the sensitivity square of variable are injected according to power grid
The L rank cumulant of electric network state variable is calculated in battle array and transfer matrix；
3) utilize Gram-Charlier series expansion, by the Probability Characteristics of electric network state variable be expressed as normal state with
The series of machine variables L order derivative composition, series coefficients are determined according to the L rank of electric network state variable standardization cumulant.
The value of the L is 5~7.
The standardization cumulant is calculated according to the following formula:
Wherein, giFor standardization after the i-th rank cumulant,For the i power of electric network state variable standard deviation, χiFor electricity
I-th rank cumulant of netted state variable, i=1 ..., L.
The calculation method of the L rank central moment is as follows:
Wherein, E is scene desired value, λiFor i-th of scene that scene is concentrated, m is the number that scene concentrates scene, βlIt is
The l rank central moment of scene, l=1 ..., L.
The step 1.4) is folded specifically includes the following steps: the Probabilistic Load Flow characteristic of scene collection is carried out probability according to the following formula
Add, obtain the trend distribution character of power grid:
Wherein, f and F is respectively the probability density function vector sum cumulative distribution function vector of electric network swim, fjAnd FjRespectively
For the probability density function vector sum cumulative distribution function vector of electric network swim under j-th of scene collection, qjFor corresponding scene collection
Probability, n are the number of scene collection.
The beneficial effects of the present invention are embodied in:
The present invention cuts down initial scene, the typical scene for classification is formed, by dividing initial scene
Class and probability are reallocated, it is possible to reduce as data fluctuations it is big and caused by Cumulants method calculate error, by each field
The characteristics of tidal flow of Jing Ji is overlapped, to obtain the overall trend distribution character of power grid.The invention enables in face of structure is complicated,
Solution in large scale, containing the electric network swim there are many enchancement factor (high proportion renewable energy, energy storage, controllable burden etc.) is asked
Topic, also ensures computational accuracy while improving solution efficiency, calculated results can be used as the reference for Electric Power Network Planning according to
According to correlative study work can be unfolded in engineering actual person accordingly.
Detailed description of the invention
Fig. 1 is initial scene distribution map in embodiment.
Fig. 2 is scene distribution map after cutting down in embodiment, and wherein 1-10 is target scene number.
Fig. 3 is the probabilistic load flow flow chart cut down based on scene.
Fig. 4 is that probability is superimposed (probability density function) schematic diagram.
Specific embodiment
The present invention is described in further details with reference to the accompanying drawings and examples.
1. the formation of the generation of scene, reduction and scene collection
1) scene generates
Firstly the need of generation database.Optimization expectational model containing stochastic variable can be convex random excellent with following form
The model of expected value of change indicates:
Random vector parameter ω in the Expection optimal time of policymaker's function of pursuit of the objective according to the actual situation, formula (2)
Following (or observing) the Discrete Stochastic number in the cards, these following random numbers clothes in the cards will be converted into
From the probability distribution of random vector parameter ω, Discrete Stochastic number here is " scene ".
In Practical Project problem, original probability estimates P by continuously or by many discrete scenes forming, and is difficult to lead to
The method for crossing parsing acquires the optimal desired value of formula (2), needs originally discrete or continuous probability measure P is further discrete
Change, i.e., with limited huge sample approximate representation probability measure P.Limited substantial amounts are obtained according to probability-distribution function
Sample approximate representation probability measure P is exactly " scene generation ".Enchancement factor in each pair of electric system completes single sample, just gives birth to
At a scene.Multiple sampling is carried out, constructs initial scene library, number in initial scene library comprising scene (or sampling time
Number) depend on research needed for precision and scene dimension.
2) scene is cut down
The number comprising enchancement factor is more and higher to required precision in usual electric system, and initial scene library is often wrapped
Containing thousands of a scenes.Enchancement factor data fluctuations bring data error will affect the accuracy of subsequent calculated result,
It is therefore desirable to carry out scene reduction to initial scene library.But so-called scene abatement feature is in this patent: 1) using quick
Forward selection procedures select typical scene (target value number is generally between 5~10), using typical scene to initial scene library into
Row subregion (classifies to initial scene), forms scene collection；2) probability reallocation is carried out to the scene collection obtained after classification,
Obtain the corresponding probability of scene collection.Based on two above feature, the present invention obtains multiple scene collection and scene collection is corresponding
Probability helps to mention compared to the degree of fluctuation for equally reducing enchancement factor with initial scene library inside each scene collection
The computational accuracy of high Cumulants method.
The known approximate model and structure for some particular problem various for modelling shown in formula (2).Wherein ζ
The mathematical notation that structure probability is estimated is as follows:
The reality of formula (3) is meant that: " solving Stochastic Optimization Model ∫ in formula (2)ΩThe expectation optimal value of f (ω, x) Pd ω "
With " solution Stochastic Optimization Model ∫ΩThe expectation optimal value of f (ω, x) Qd ω " is of equal value.Therefore the P scale in formula (2) compared with
When greatly and being difficult to be fully described, probability measure Q solving model can be simplified using the approximation of P." how to obtain optimal letter
Changing scene collection Q " is exactly described " scene reduction ".It is inherently an optimization problem about Q that scene, which cuts down problem,.
The reduction fast speed of fast forword back-and-forth method is cut down the intuitive therefore of the invention scene of thought and is cut down in algorithm
Using the method.In order to illustrate fast forword back-and-forth method, need to introduce optimal reduction problem:
Wherein, N is initial scene number, and n is the target scene number after cutting down, c (ωi,ωj) it is scene ωiAnd ωjBetween
Euclidean distance, piFor scene ωiThe probability of generation.Formula (4) illustrates that initial scene set { 1,2 ..., N } is divided into
Two parts, a part are the scene number set J being cut in, another part be remain scene number set 1,
2 ..., N } J, in the optimization problem, objective function is the smallest DJValue, optimized variable is J.
There is no effective algorithms for the problems in usual situation following formula (4), but for the scene number #J=N-1 being cut in
This case, the solution of the problem become relatively easy.Fast forword back-and-forth method is to be equivalent to choosing at this time based on this case
A point u is selected, is cut in it not, formula (4) conversion are as follows:
Wherein, u is the point for not needing to delete, and in addition to this all points are all deleted, and just has selected one so least
Then the scene that can be cut in continues similarly with the selection for considering other scenes and probability redistribution problem again.Fast forword
The step of back-and-forth method, is as follows:
1. initial calculation
It enablesFirst is selected
The point for not needing to delete comeJ1={ 1,2 ..., N } { u1}。
2. cycle calculations
Determining i-th (i > 1) the point u for not needing to delete being selectediDuring,Calculation method it is different
In initial calculation, need to use the value in (i-1)-th calculating, i.e.,It is (i-1)-th
A point for not needing to delete chosen,I-th of point being selectedJi=Ji-1\{ui}。
3. probability is reallocated
It is substantially exactly to classify to N number of scene in initial scene library that probability, which is reallocated,.If oi(i=1 ..., N) be
Any scene in initial scene library, pi(i=1 ..., N) is the probability that any scene in initial scene library occurs, sj(j=
1 ..., n) it is any scene (typical scene) retained in scene library.The present invention is according between initial scene and typical scene
Probability metrics classify to initial scene, i.e., for any one initial scene, calculate separately the initial scene and each allusion quotation
Probability metrics between type scene；Corresponding typical scene is divided into one when by the initial scene and acquirement probability metrics minimum value
Class, i.e., i-th of scene will be divided into scene collection in initial scene libraryIt is denoted as i ∈ j,
pic(oi,sj) it is scene oiAnd sjBetween probability metrics.Any scene collection setjThe probability q of appearancejIt is obtained by following formula:
3) formation of scene collection
Present invention definition converges to typical scene sjAll initial scenes (including scene sjItself) constitute scene collection
setj, setjIn include initial scene number be numj, then have:
setj={ oi|i∈j} (7)
The classification to initial scene is completed in this way, that is, forms scene collection.
2. the probabilistic load flow cut down based on scene
Cumulants method is very sensitive to the fluctuation of data, therefore a high proportion of enchancement factor bring data in power grid
Biggish fluctuation and randomness will lead to error calculated increase.This is because carrying out Taylor expansion at benchmark operating point
2 times or more high-order terms are had ignored when linearisation, if enchancement factor fluctuation is larger, then the error linearized can also increase therewith
Greatly.The present invention has carried out division operation to initial scene library, and the boundary between scene sample has been determined using scene classification method,
The subregion fluctuation of enchancement factor being limited in where it, i.e., inside scene collection.This equally reduces the fluctuation of enchancement factor
Degree, to improve the computational accuracy of Cumulants method.The Cumulants method used below to the present invention is illustrated.
1) scene collection (node injection variable) cumulant calculates
The number of scenes for including in initial scene library is N, and scene integrates number as n, and the probability after fast forword selection divides again
The classification number to scene N number of in initial scene library is completed with process, forms multiple scene collection.Institute is concentrated for research scene
Some scene samples calculate each rank central moment for obtaining scene using the method for statistics:
Wherein, E is scene desired value, λiFor i-th of scene that research scene is concentrated, m is that research scene concentrates scene
Number, βlIt is the l rank central moment of scene.
Relationship between central moment and cumulant is shown below:
Wherein, γlFor l rank cumulant.The present invention takes preceding 7 rank cumulant, and precision is met the requirements.According to formula
(9)-(11) obtain the preceding 7 rank cumulant of research scene collection, are used for subsequent calculating.
2) Cumulants method
In order to which application of the Cumulants method in probabilistic load flow is better described, it is firstly introduced into electric system linearisation
Thought.The form of the modal equation of electric system and branch equation matrix is indicated, and in benchmark operating point (i.e. in scene collection
Typical scene) at Taylor series expansion is carried out to it, ignore 2 times or more high-order terms:
In view of meeting at benchmark operating point:
Benchmark operating point can be obtained to the lienarized equation of random perturbation Δ W:
In formula: subscript 0 indicates that benchmark operating point, W are that node injects variable, and X is node state variable, and Z is membership
Variable.S0For sensitivity matrix, T0For transfer matrix.
There are two critical natures for cumulant tool:
1. additive property: each rank cumulant of the sum of independent random variable be equal to each stochastic variable each rank cumulant it
With.
2. homogeneity: the r rank half that a times of stochastic variable of r (r >=1, r are integers) rank cumulant is equal to the variable is constant
The a of amountrTimes.
Utilize node injection rate shown in this two critical natures and formula (14) and node state amount and line status
Relationship between amount, can be obtained according to the cumulant of node injection rate node state amount and line status amount partly not
Variable.
The random change is determined by each rank cumulant of stochastic variable (such as the state variables such as node voltage, line power)
There are many kinds of the methods for measuring probability characteristics.Gram-Charlier series expansion has stronger representativeness, series expansion mode
It is as follows:
Wherein, N (t) is normpdf, and t is the stochastic variable after standardization, χiFor stochastic variable
I-th rank cumulant, giFor the i-th rank cumulant after standardization, HiIt (t) is the i-th rank Hermite multinomial, EX is random becomes
Measure the desired value of X, δXFor the standard deviation of stochastic variable X,For the i power of stochastic variable X standard deviation.
3) combination of scene reduction and Cumulants method
By scene reduction and probability reallocation link, multiple scene collection set are obtainedjAnd its corresponding probability qj；It is each
Cumulant inside scene collection can also be obtained by the method for statistics.Each scene collection has using Cumulants method
Carry out all key elements of probabilistic load flow.The Cumulants method key step cut down based on scene is as shown in Figure 3:
1. carrying out scene reduction, the probability that typical scene, corresponding scene collection and scene collection occur is obtained, using statistics
Method calculates each rank cumulant obtained under scene collection.
2. being directed to scene collection setj, since it has whole key elements needed for Cumulants method, using cumulant
Method carries out probabilistic load flow, obtains the Probability Characteristics of the scene collected state amount, the probability density letter comprising quantity of state
Number fj(x) and cumulative distribution function Fj(x)。
3. calculating the probability density function and probability-distribution function of certain quantity of state (i.e. state variable) under each scene collection, lead to
Cross the Probability Characteristics (Fig. 4) that probability principle of stacking obtains the quantity of state:
All quantity of states (including node voltage amplitude, the active power of phase angle and branch road, reactive power) are pressed
Illuminated (18) calculates, i.e., is overlapped to the characteristics of tidal flow of each scene collection, then obtains the overall trend distribution character of power grid.
Simulation example
1. generating database
Acquired initial data, such as main distribution network structure, enchancement factor characteristic, system initial parameter are arranged, and
Sliding-model control is carried out to data according to its probability-distribution function, according to the distribution characteristics of enchancement factor in electric system to wind
These enchancement factors such as machine, photovoltaic, energy storage device, controllable burden are sampled, and every completion single sample just obtains a scene.
Multiple sampling is carried out, great amount of samples is obtained, generates initial scene library.By taking number of scenes is 500 initial scene library as an example, initially
Scene distribution represents 500 initial scenes as shown in Figure 1, comprising 500 sample points in Fig. 1, for convenience on plan view
It is shown and is illustrated, each scene only includes 2 factors；The value range of each factor is that the integer between 0-100 (includes
0 and 100)；The probability that each scene occurs is identical, is all 0.2%, and the sum of probability is 1.
2. pair initial scene library generated carries out scene reduction
A large amount of original scenes are cut down with quick former generation method.It is eliminated using quick former generation and field is carried out to initial scene library
Scape is cut down, and specific process of cutting down is referring to formula (2)-(6), and target scene number is 10, scene distribution such as Fig. 2 institute after reduction
Show, it can be seen that typical scene is multiple cluster centres of initial scene library, the field in initial scene library, around typical scene
Scape distribution is more intensive.
3. forming scene collection
Classified according to formula (7), (8) to initial scene, forms scene collection.
4. probabilistic load flow
In view of application background be containing blower, photovoltaic, energy storage device, controllable burden these enchancement factors power grid, this hair
The bright calculating for carrying out Probabilistic Load Flow to power grid using Cumulants method.
1) cumulant that scene collection is calculated according to formula (9)-(11), establishes the connection between scene collection and cumulant
System.
2) by electric system modal equation and branch equation Taylor expansion is carried out at typical scene, ignore it is secondary and
Above high-order term obtains the lienarized equation of formula (12), (14) form, to calculate the cumulant of state variable.
3) Gram-Charlier series is utilized, the Probability Characteristics of state variable are expressed as normal state according to formula (15)
The series of stochastic variable all-order derivative composition determines series coefficients according to each rank of state variable standardization cumulant.
5. scene is cut down and the combination of Cumulants method
By the Probabilistic Load Flow characteristic comprehensive analysis of the probability of scene collection and each scene collection, obtain being investigated according to formula (18)
The probability density characteristics of electric network state amount.
Simulation result shows using method of the invention, compared with conventional Cumulants method, realizes initial scene library
Division operation can equally reduce the degree of fluctuation of enchancement factor, to improve Cumulants method computational accuracy.
In short, the present invention proposes the concept of scene collection, after realizing that scene is cut down, the classification to initial scene is completed,
The determination sharpening for making boundary between scene in this way, helps to combine with cumulant calculating power system load flow, to larger
Initial scene library cut down and obtain scene collection on the basis of, carry out probability with Cumulants method inside each scene collection
While improving efficiency, computational accuracy can be effectively ensured in the calculating of trend.Method proposed in the present invention can be applied to
The Load flow calculation of the large-scale complex power grid of the renewable energy containing high proportion can be solved effectively due to electric system enchancement factor wave
Cumulants method calculation of tidal current accuracy decline problem caused by moving greatly, it is ensured that the height of the probabilistic load flow of power grid is quasi-
True property.
Claims (8)
1. a kind of Probabilistic Load calculation method cut down based on scene, it is characterised in that: the following steps are included:
1.1) enchancement factor in electric system is sampled, obtains initial scene library；
1.2) initial scene library is cut down, obtains typical scene；According to typical scene to whole fields in initial scene library
Scape is classified, and scene collection is obtained, and the general of scene collection is calculated according to the scene classification of initial scene library and probability characteristics
Rate feature；
1.3) probabilistic load flow is carried out to each scene collection using the Probabilistic Load Flow algorithm based on Cumulants method respectively；
The step 1.3) specifically includes the following steps:
1) by electric system modal equation and branch equation Taylor expansion is carried out at typical scene, then by linear
Change, obtains sensitivity matrix and transfer matrix；
2) all scenes are concentrated for scene, the L rank central moment of scene are calculated using the method for statistics, according in the L rank
Heart square calculates the L rank cumulant of power grid injection variable, according to power grid inject the L rank cumulant of variable, sensitivity matrix and
The L rank cumulant of electric network state variable is calculated in transfer matrix；
3) Gram-Charlier series expansion is utilized, the Probability Characteristics of electric network state variable are expressed as normal state and are become at random
The series of L order derivative composition is measured, series coefficients are determined according to the L rank of electric network state variable standardization cumulant；
1.4) probability characteristics for combining scene collection carry out probability superposition to the characteristics of tidal flow of each scene collection, obtain the tide of power grid
Flow distribution characteristic.
2. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The step 1.1) is specifically includes the following steps: obtain the feature distribution of enchancement factor according to the initial data of electric system, so
Afterwards using Monte Carlo Method to enchancement factor carry out n times sampling, every progresss single sample just obtain an initial scene and its generally
Rate feature.
3. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The step 1.2) specifically includes the following steps:
1) initial scene is cut down with fast forword back-and-forth method, until the initial scene retained reaches the target scene of setting
Number n, using the n of reservation initial scenes as typical scene；
2) classified according to the probability metrics between initial scene and typical scene to initial scene, initial fields are made by classification
Scape is converged according to typical scene, so that it is determined that the boundary between initial scene；
3) the probability q occurred for j-th of scene collectionjIt is obtained by following formula:
Wherein, i ∈ j indicates that i-th of scene will be divided into j-th of scene collection, p in initial scene libraryiIt is in initial scene library
The probability that i scene occurs.
4. a kind of Probabilistic Load calculation method cut down based on scene according to claim 3, it is characterised in that:
To Mr. Yu's initial fields scape oi, calculate separately the initial scene and each typical scene sjBetween probability metrics；By the initial scene
It is divided into one kind with the minimum corresponding typical scene of probability metrics, i.e., i-th of scene will be divided into scene collection in initial scene library
setj,pic(oi,sj) it is scene oiAnd sjBetween probability metrics, c (oi,sj) it is scene oi
And sjBetween Euclidean distance.
5. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The value of the L is 5~7.
6. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The standardization cumulant is calculated according to the following formula:
Wherein, giFor standardization after the i-th rank cumulant,For the i power of electric network state variable standard deviation, χiFor power grid shape
I-th rank cumulant of state variable, i=1 ..., L.
7. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The calculation method of the L rank central moment is as follows:
Wherein, E is scene desired value, λiFor i-th of scene that scene is concentrated, m is the number that scene concentrates scene, βlIt is scene
L rank central moment, l=1 ..., L.
8. a kind of Probabilistic Load calculation method cut down based on scene according to claim 1, it is characterised in that:
The step 1.4) obtains specifically includes the following steps: the Probabilistic Load Flow characteristic of scene collection is carried out probability superposition according to the following formula
The trend distribution character of power grid:
Wherein, f and F is respectively the probability density function vector sum cumulative distribution function vector of electric network swim, fjAnd FjRespectively
The probability density function vector sum cumulative distribution function vector of electric network swim, q under j scene collectionjFor the probability of corresponding scene collection,
N is the number of scene collection, and x is state variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710552406.4A CN107276093B (en) | 2017-07-07 | 2017-07-07 | Power system probability load flow calculation method based on scene reduction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710552406.4A CN107276093B (en) | 2017-07-07 | 2017-07-07 | Power system probability load flow calculation method based on scene reduction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107276093A CN107276093A (en) | 2017-10-20 |
| CN107276093B true CN107276093B (en) | 2019-10-18 |
Family
ID=60073193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710552406.4A Active CN107276093B (en) | 2017-07-07 | 2017-07-07 | Power system probability load flow calculation method based on scene reduction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107276093B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109038689B (en) * | 2018-09-13 | 2021-11-05 | 南方电网科学研究院有限责任公司 | Ultra-short-term optimal scheduling method for power system |
| CN112508402B (en) * | 2020-12-04 | 2024-02-27 | 国网湖北省电力有限公司电力科学研究院 | DC power transmission curve scene generation method, electronic equipment and medium |
| CN112688318B (en) * | 2020-12-21 | 2022-10-14 | 国网浙江省电力有限公司电力科学研究院 | Method for constructing new energy station internal frequency modulation instruction decomposition model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104682387A (en) * | 2015-03-10 | 2015-06-03 | 东南大学 | Probability load flow calculation method based on multi-zone interactive iteration |
| CN105322535A (en) * | 2015-10-13 | 2016-02-10 | 河海大学 | Two-stage optimal power flow calculation method for power supply containing unified power flow controller |
| CN105529714A (en) * | 2016-02-17 | 2016-04-27 | 中国农业大学 | Normal distribution combination characteristic-based rapid probabilistic power flow calculation method |
| CN105656031A (en) * | 2016-02-17 | 2016-06-08 | 中国农业大学 | Security risk assessment method of wind-power-included electric power system based on Gaussian mixture distribution characteristics |
-
2017
- 2017-07-07 CN CN201710552406.4A patent/CN107276093B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104682387A (en) * | 2015-03-10 | 2015-06-03 | 东南大学 | Probability load flow calculation method based on multi-zone interactive iteration |
| CN105322535A (en) * | 2015-10-13 | 2016-02-10 | 河海大学 | Two-stage optimal power flow calculation method for power supply containing unified power flow controller |
| CN105529714A (en) * | 2016-02-17 | 2016-04-27 | 中国农业大学 | Normal distribution combination characteristic-based rapid probabilistic power flow calculation method |
| CN105656031A (en) * | 2016-02-17 | 2016-06-08 | 中国农业大学 | Security risk assessment method of wind-power-included electric power system based on Gaussian mixture distribution characteristics |
Non-Patent Citations (1)
| Title |
|---|
| 采用分段离散化和高斯混合模型的多场景概率潮流计算;叶林等;《电力系统自动化》;20170625;第41卷(第12期);第131～137页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107276093A (en) | 2017-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | A missing power data filling method based on improved random forest algorithm | |
| CN106446540B (en) | A kind of Wind turbines health status real time evaluating method | |
| Singh et al. | An efficient technique for reliability analysis of power systems including time dependent sources | |
| CN107104442B (en) | Method for calculating probability load flow of power system including wind power plant by considering parameter ambiguity | |
| CN107039995A (en) | A kind of distributed photovoltaic power access planing method for considering power quality problem | |
| CN108599172B (en) | Transmission and distribution network global load flow calculation method based on artificial neural network | |
| CN111401755A (en) | Multi-new-energy output scene generation method, device and system based on Markov chain | |
| CN111103477B (en) | Electrical load monitoring method and system based on 3D convolutional neural network | |
| CN107276093B (en) | Power system probability load flow calculation method based on scene reduction | |
| CN102832617A (en) | Large power grid transient state stabilization analyzing method based on precision pattern discrimination | |
| CN103455948B (en) | A kind of distribution system multi-dimensional multi-resolution Modeling and the method for analysis | |
| Wang et al. | Distribution network state estimation based on attention-enhanced recurrent neural network pseudo-measurement modeling | |
| Xiao et al. | Optimal sizing and siting of soft open point for improving the three phase unbalance of the distribution network | |
| CN106503279A (en) | A kind of modeling method for transient stability evaluation in power system | |
| CN112865089A (en) | Improved large-scale scene analysis method for active power distribution network | |
| CN113890042B (en) | Load flow calculation method based on active power distribution network | |
| CN103353895A (en) | Pre-processing method of power distribution network line loss data | |
| CN105896547A (en) | Large power network graded voltage control method under wind power access | |
| CN108694475B (en) | Short-time-scale photovoltaic cell power generation capacity prediction method based on hybrid model | |
| CN105914752A (en) | Pilot node selection method based on clustering by fast search and density peaks | |
| CN111861256B (en) | Active power distribution network reconstruction decision method and system | |
| CN106058862B (en) | It is a kind of meter and voltage stability power distribution network method for fast reconstruction | |
| CN117477544A (en) | LSTM ultra-short-term photovoltaic power prediction method and system integrating time mode features | |
| CN115173421A (en) | Probability optimal power flow calculation method based on progressive Latin hypercube sampling | |
| Habib et al. | Revolutionizing Wind Power Prediction—The Future of Energy Forecasting with Advanced Deep Learning and Strategic Feature Engineering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210830 Address after: 510663 3 building, 3, 4, 5 and J1 building, 11 building, No. 11, Ke Xiang Road, Luogang District Science City, Guangzhou, Guangdong. Patentee after: China South Power Grid International Co.,Ltd. Patentee after: XI'AN JIAOTONG University Address before: 510623 Guangdong city of Guangzhou province Tianhe District Pearl River Metro Chinese Sui Road No. 6 Patentee before: POWER GRID TECHNOLOGY RESEARCH CENTER. CHINA SOUTHERN POWER GRID Patentee before: XI'AN JIAOTONG University Patentee before: China South Power Grid International Co.,Ltd. |